Prevalence and Risk Factors in HIV-infected Persons for Echocardiographic Abnormalities in the Era of Modern HAART

The natural history of HIV infection has been associated with a diverse array of adverse metabolic complications that may increase the risk of developing cardiovascular disease and accelerate risk of other diseases.The Study to Understand the Natural History of HIV/AIDS in the Era of Effective Therapy

(‘SUN’ Study) is a prospective cohort study designed to examine risk factors for complications of treating HIV infection and longer survival.Patients are followed from 7 clinics in 4 U.S. cities.The primary objectives of this analysis were to characterize baseline cardiac function of subjects by modern echocardiographic methods and to determine clinical, behavioral, and laboratory predictors of prevalent echocardiographic abnormalities.

#978

K Mondy1, J Gottdiener2, ET Overton1, K Henry3, L Conley4,T Bush4, J Hammer5, C Carpenter6, EM Kojic6, JT Brooks4, and SUN Study Investigators

1Washington Univ Sch of Med, St Louis, MO, US; 2 Univ of Maryland, Baltimore, Maryland, US; 3HIV Program, Hennepin County Med Ctr, Univ of Minnesota, MN, US; 4CDC, Atlanta, GA, US; 5Denver Infectious Disease Consultants , Denver, US; 6The Miriam Hosp, Providence, RI, US

The findings and conclusions in this poster presentation are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.

Background

Methods

Results: Echocardiographic Characteristics of Subjects (Table 1)

SUN Study Subjects: Age >18, either ARV-naïve with CD4 cell count 100-500 cells/mm3 or HAART-experienced with CD4 cell count >100 cells/mm3; no immunomodulating therapy or AIDS-defining illness in 60 days preceding baseline enrollment. Methods/Data Collection: Baseline resting echocardiography with tissue doppler imaging, carotid intima media thickness, patient demographics, behavioral risk factors, and fasting laboratory data were analyzed for the SUN cohort (n=667, tables 1-2); subgroup analyses were performed for the following echocardiographic abnormalities: left ventricular systolic dysfunction, diastolic dysfunction, pulmonary hypertension, and left ventricular hypertrophy. Sources of data included anonymous patient interview and laboratory data obtained at baseline study visit as well as abstraction of data from outpatient medical records. Echocardiographic and carotid data were read at central reading sites (U. of Maryland, Baltimore and U. of Southern California, respectively).Statistics: Chi-square or Fisher’s exact tests were used for categorical variables. Continuous variables were compared using the Student’s t-test or Mann-Whitney U test. Multivariate logistic regression were used to identify independent predictors of echocardiographic outcomes from among factors significant in univariate analyses. All p values were two-tailed. Analysis was performed using SAS software version 9.1.

Results: Patient Characteristics (Table 2)

In the era of modern HAART, subtle abnormalities in cardiac function were prevalent within this patient cohort.

Echocardiographic abnormalities were predicted not only by traditional cardiac risk factors but also modifiable risk factors including illicit drug use.

Of the ARV therapy evaluated, the only associations were current use of ritonavir-boosted PIs with pulmonary hypertension and current AZT use with left ventricular hypertrophy.

The current analysis suggests lifestyle modification should be a high priority in the management of chronic HIV disease.

Further longitudinal assessment of cardiac function is anticipated in this HIV-infected cohort to determine whether these cardiac abnormalities remain stable over time.

Results (Multivariate Analysis) : Predictors of Echocardiographic Abnormalities (Table 3)

Conclusions

ABSTRACT (updated)

Background: Cardiac function among HIV-infected persons in the contemporary treatment era has not been well characterized.

Methods: The SUN Study is a prospective cohort of 682 HIV-infected patients receiving care at clinics in Denver, Minneapolis, Providence, and St. Louis. At baseline, all patients underwent standardized echocardiographic examination. Using multivariate logistic regression modeling, we identified independent predictors of left ventricular systolic dysfunction (LVSD = ejection fraction <55%), diastolic dysfunction (DD), pulmonary hypertension (PHTN = right ventricular systolic pressure >30 mmHg), and LV hypertrophy (LVH = LV mass >45 (women) and >49 (men) g/m2).

Results: Characteristics for 643 subjects with available data were: median age 41 years, 23% women, 27% Black, mean body mass index (BMI) 26.4 kg/m2, 77% on HAART, mean CD4 cell count (CD4) 519 cell/µL, 54% with HIV viral load <400 copies/mL, 14% hypertensive, 6% diabetic, 44% smokers. Eighteen percent of 640 evaluable subjects had LVSD (111 [18%] mild, 3 [<1%] moderate, 0 severe), 25% of 660 evaluable subjects had DD (96 [15%] grade 1 [G1], 9 [<1%] G2, 59 [9%] G3), 22% of 316 evaluable patients (i.e., had any tricuspid regurgitant flow present) had Pulm HTN (51 [16%] mild, 16 [<1%] moderate, 5 [<1%] severe), 6.6% of 604 evaluable subjects had LVH (24 [4%] mild, 8 [1%] each moderate and severe). Multivariate analyses demonstrated the following independent associations: for LVSD – male gender (odds ratio [OR] 2.1), carotid intima-media thickness [IMT] (OR 1.6 if >0.71 mm [median value]), and smoking (OR 1.5); for DD – HTN (OR 2.1), and use of inhaled nitrites (OR 0.6); for Pulm HTN – Total Cholesterol (OR 2.6 if >154mg/dL [1st quartile]), Age > 35 years (O.R. 2.2), and current use of a boosted PI (O.R. 1.9); and for LVH – body mass index (OR 3.5 if >25.4 [median value]), use of marijuana (OR 2.0) IMT (OR 1.9 if >0.79 mm [upper quartile]), and current AZT use (OR 1.5) (p<0.05 for all).

Conclusions: In this cohort of contemporary HAART-era patients, subclinical abnormalities in cardiac function were detected frequently. Functional abnormalities were mostly associated with expected and often modifiable risks. Our analysis suggests lifestyle modification should become a greater priority in the management of chronic HIV disease.

Providence

Minneapolis

St. LouisDenver

Characteristic No. (%)

(N=643)

Age (mean + SEM) 41.0 + 0.4

Gender

Men

Women

495 (77)

148 (23)

Race

White, non-Hispanic

Black, non-Hispanic

Hispanic

Unknown

388 (61)

173 (27)

64 (10)

12 ( 2)

Years Since HIV Diagnosis (mean + SEM) 6.0 + 0.2

History of Opportunistic Infection 121 (19)

Co-infection with Hepatitis B 21 (5.0)

Co-infection with Hepatitis C 70 (13)

Current Smoker 282 (44)

Current Drug Use

Cocaine

Marijuana

IVDU

Inhaled nitrites

63 (10)

160 (25)

7 (1)

107 (17)

On HAART

HIV RNA<400 copies

493 (77)

350 (71)

Type of HAART*

Ritonavir-boosted PI

Unboosted PI

NNRTI

3 or more NRTIs

183 (29)

50 (8)

237 (38)

112 (18)

Time (years) on HAART (mean ± SEM)

NRTIs

PIs

NNRTIs

3.0 + 0.1

1.6 + 0.9

1.4 + 0.8

CD4 Count (mean cells/mm3 + SEM)

current

nadir

519 + 11.1

221 + 6.5

BMI (mean kg/m2 ± SEM) 26.4 + 0.2

Fasting Glucose (mean mg/dL + SEM) 95 + 1.1

Fasting Total Cholesterol (mean mg/dL + SEM) 185 + 2.0

Fasting LDL (mean mg/dL + SEM) 109 + 1.6

Fasting HDL (mean mg/dL + SEM) 43 + 0.6

Fasting TG (mean mg/dL + SEM) 178 + 5.5

Current Use of Lipid-lowering Therapy 72 (11)

Diagnosis of Hypertension 92 (14)

On Antihypertensive Therapy 107 (17)

Diagnosis of Diabetes 38 (6)

History of Myocardial Infarction 2 (<1)

Mean Carotid IMT (mm + SEM)

Median Carotid IMT

0.73 + 0.01

0.71

Acknowledgements: We wish to gratefully acknowledge all SUN study investigators and

their support staff, Cerner Corporation, and all of the study participants that have devoted

their time and effort to this research endeavor.

Significant Predictors of Left Ventricular Systolic Dysfunction (EF<55%)

Predictor Odds Ratio 95% C. I. P value

Gender male 2.05 1.19 – 3.71 0.013

Carotid IMT > 0.71 mm (median) 1.64 1.08 – 2.51 0.022

Current smoking 1.54 1.02 – 2.35 0.041

Significant Predictors of Diastolic Dysfunction (Grade 1-3)

Predictor Odds Ratio 95% C. I. P value

Hypertension 2.06 1.32-3.19   0.001

Poppers in last 30 days 0.56 0.35 – 0.96 0.034

Significant Predictors of Pulmonary Hypertension

Predictor Odds Ratio 95% C. I. P value

Total Cholesterol >154 mg/dL (lower quartile) 2.58 1.37 – 5.09 0.004

Age >35 years (lower quartile) 2.22 1.23 – 4.03 0.008

Boosted PI 1.90 1.12 – 3.29 0.019

Significant Predictors of Left Ventricular Hypertrophy

Predictor Odds Ratio 95% C. I. P value

BMI >25.4 (median) 3.53 2.30 – 5.52 <0.001

Marijuana in last 6 months 2.01 1.35 – 3.03 <0.001

Carotid IMT >0.79 mm (upper quartile) 1.90 1.24 – 2.93 0.003

Current AZT use 1.54 1.04 – 2.27 0.030

Significant Predictors of Left Atrial Enlargement

Predictor Odds Ratio 95% C. I. P value

Hepatitis C 2.27 1.31 – 4.06 0.004

Marijuana in last 6 months 1.74 1.17 – 2.60 0.006

Age >46 years (upper quartile) 1.73 1.13 – 2.67 0.012

VAT/SAT < 0.58 (median) 1.49 1.03 – 2.17 0.036

*Pulmonary hypertension could only be evaluated in those subjects with the presence of tricuspid regurgitant flow.S.E.M. = standard error of the mean value

Characteristic No. (%) Mean + S.E.M. Median (Range)

Systolic function (LVEF)

Normal: >55% 523 (82) 61.0 + 0.2 60.2 (55.0 – 77.1)

Mild: 45-55% 114 (18) 53.3 + 0.1 53.6 (45.4 – 55.0)

Moderate: 30-45% 3 ( <1) 39.1 + 3.9 42.7 (31.4 – 43.2)

Severe: <30% 0 (0.0) N/A N/A

Diastolic function

Grade 0: normal 470 (74) N/A N/A

Grade 1: impaired relaxation 96 (15) N/A N/A

Grade 2: pseudonormal 9 ( <1) N/A N/A

Grade 3: restrictive 59 (9) N/A N/A

Pulmonary HTN (RVP mmHg)

Normal: < 31 136 (43) 27.1 + 0.3 27.5 (14.4 – 31.0)

Borderline 31- 35 108 (34) 33.1 + 0.1 33 (31.1 – 36.0)

Mild: 36-40 51 ( 16) 38.2 + 0.2 37.9 (36.1 – 40.9)

Moderate: 41-50 16 ( <1) 44.6 + 0.7 44.8 (41.1 – 49.9)

Severe: >50 5 ( <1) 67.8 + 7.5 57.1 (54.8 – 90.3)

Left ventricular mass (g/m2.7)

Normal: (<45 W; <49 M) 564 (93) 32.8 + 0.3 32.6 (10.2 – 48.8)

Mild abnormal: (45-51 W; 49-55 M)

24 (4) 49.2 + 0.5 49.1 (45.3 – 55.9)

Moderate: (52-58 W; 56-63 M) 8 ( 1) 58.0 + 1.1 57.1 (54.0 – 62.8)

Severe abnormal: (>59 W; >64 M) 8 ( 1) 83.3 + 10.7 68.7 (60.7 – 151.4)

Left atrial volume (ml/m2 BSA)

Normal: <29 360 (61) 23.3 + 0.2 23.6 (11.0 – 29.0)

Mild abnormal: 29-33 121 (21) 31.4 + 0.1 31.4 (29.1 – 34.0)

Moderate abnormal: 34-39 64 ( 11) 36.4 + 0.2 36.4 (34.2 – 39.9)

Severe abnormal: >40 45 ( 8) 45.1 + 0.9 43.3 (40.1 – 71.2)

*Some patients were receiving both PIs and NNRTIs.